Phylogeographic Origin of Helicobacter pylori Determines HostAdaptive Responses upon Coculture with Gastric Epithelial Cells

Alexander Sheh,a Rupesh Chaturvedi,b,e D. Scott Merrell,f Pelayo Correa,b Keith T. Wilson,b,c,d,e James G. Foxa,g
Division of Comparative Medicine, Massachusetts Institute of Technology, Cambridge, Massachusetts, USAa; Division of Gastroenterology, Department of Medicine,b Department of Cancer Biology,c and Department of Pathology, Microbiology and Immunology,d Vanderbilt University School of Medicine, Nashville, Tennessee, USA; Veterans Affairs Tennessee Valley Healthcare System, Nashville, Tennessee, USAe; Department of Microbiology and Immunology, Uniformed Services University of the Health Sciences, Bethesda, Maryland, USAf; Department of Biological Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, USAg

While Helicobacter pylori infects over 50% of the worlds population, the mechanisms involved in the development of gastric disease are not fully understood. Bacterial, host, and environmental factors play a role in disease outcome. To investigate the role of bacterial factors in H. pylori pathogenesis, global gene expression of six H. pylori isolates was analyzed during coculture with gastric epithelial cells. Clustering analysis of six Colombian clinical isolates from a region with low gastric cancer risk and a region with high gastric cancer risk segregated strains based on their phylogeographic origin. One hundred forty-six genes had increased expression in European strains, while 350 genes had increased expression in African strains. Differential expression was observed in genes associated with motility, pathogenicity, and other adaptations to the host environment. European strains had greater expression of the virulence factors cagA, vacA, and babB and were associated with increased gastric histologic lesions in patients. In AGS cells, European strains promoted signicantly higher interleukin-8 (IL-8) expression than did African strains. African strains signicantly induced apoptosis, whereas only one European strain signicantly induced apoptosis. Our data suggest that gene expression proles of clinical isolates can discriminate strains by phylogeographic origin and that these proles are associated with changes in expression of the proinammatory and protumorigenic cytokine IL-8 and levels of apoptosis in host epithelial cells. These ndings support the hypothesis that bacterial factors determined by the phylogeographic origin of H. pylori strains may promote increased gastric disease.

elicobacter pylori infects over 50% of the worlds population and has been associated with the development of gastritis, ulcers, and gastric cancer (1). In spite of its high prevalence, chronic H. pylori infection leads to clinical symptoms in approximately 20% of infected individuals, with only 1 to 2% developing gastric adenocarcinoma. The mechanisms evoked in H. pylori pathogenesis remain incompletely characterized but suggest that host, environmental, and bacterial factors determine the outcome of H. pylori infection. Among the bacterial factors associated with H. pylori-associated disease, genes utilized for motility/chemotaxis(24), acid acclimation(56), adherence to the mucosa (7), and damaging host epithelial cells(89) are necessary for H. pylori colonization of the gastric mucosa and have been associated with increased risk of clinically relevant gastric disease (1). The absence of these virulence factors results in reduced pathogenicity or inability of H. pylori to colonize (24, 67). The best-known virulence factor in H. pylori is the cytotoxin-associated gene A (CagA) (1011). CagA is translocated into the host cells via a type IV secretion system (T4SS), whereby it becomes phosphorylated by tyrosine kinases. CagA modulates eukaryotic signaling networks, leading to hyperproliferation, inammation, apoptosis, and cancer (1214). However, current determination of the potential virulence of H. pylori strains is performed by PCR genotyping of virulence factors such as cagA, vacA, and babA (1, 9, 15) and does not account for bacterial gene expression levels, which may affect the pathological outcome in the host (16). Previous microarray studies, both in vitro and in vivo, have focused on the effects of the environment or the host on strains of H. pylori and their isogenic mutants (1723). The current study is the rst to compare expression proles of multiple clinical isolates of H. pylori interacting with host cells.

Variation between H. pylori strains may partly account for differences in gastric cancer incidence rates in different locations worldwide. In Colombia, there is a greater than 90% prevalence of H. pylori infection, but geographically distinct areas differ greatly in gastric cancer incidence (2425). Colombian populations in the high Andes have a higher incidence of gastric cancer associated with H. pylori, while coastal populations have a reduced incidence of gastric cancer (25). Multilocus sequence typing (MLST) analysis of Colombian H. pylori strains from these regions have revealed an association between strains from high-risk areas and ancestral European strains of H. pylori, while strains from low-risk regions are mostly associated phylogenetically with African strains (26). However, the mechanisms by which these genetic changes mediate greater virulence have not been characterized. In the present study, six clinical isolates from regions with low and high incidence of gastric cancer were analyzed using an H. pylori microarray to determine differences in expression associated with increased gastric cancer risk. Comparison of the H. pylori strains based on phylogeographic origin identied 496 genes

Received 25 October 2012 Returned for modication 24 November 2012 Accepted 21 April 2013 Published ahead of print 29 April 2013 Editor: S. M. Payne Address correspondence to James G. Fox, jgfox@mit.edu. Supplemental material for this article may be found at http://dx.doi.org/10.1128 /IAI.01182-12. Copyright 2013, American Society for Microbiology. All Rights Reserved. doi:10.1128/IAI.01182-12

2468

iai.asm.org

Infection and Immunity

p. 2468 2477

July 2013 Volume 81 Number 7

Origin Alters Transcription and Virulence of H. pylori

that were differentially expressed between European and African strains. While all strains in the study are cagA and vacA s1m1, there is an upregulation of key virulence factors (cagA and vacA) in European strains. Expression differences were also observed in genes associated with bacterial survival within the host, such as acid acclimation, detoxication, and motility genes. Analysis of interleukin-8 (IL-8) expression and apoptosis in gastric epithelial cells cocultured with the isolates demonstrated that European strains increase host IL-8 expression while reducing induction of apoptosis. These ndings support histopathologic data describing more-severe gastric lesions in patients infected with European strains than in patients infected with African strains. These data suggest that differential expression of bacterial factors greatly affect host-pathogen interactions and may play an important role in carcinogenesis.
MATERIALS AND METHODS
Bacteria and cells. Helicobacter pylori isolates PZ5004, PZ5024, PZ5026, PZ5056, PZ5080, and PZ5086 were cultured on blood agar (tryptic soy agar [TSA] with sheep blood; Remel, Lenexa, KS) or brucella broth with 5% fetal bovine serum under microaerobic conditions (10% H2, 10% CO2, 80% N2). Bacteria for microarray and motility experiments were collected after 18 h of growth in liquid culture. Growth curve experiments demonstrated that all six strains were in log phase and active at this time point (see Fig. S1 in the supplemental material). These cagA vacA s1m1 strains were isolated from Colombian patients (ages ranging from 47 to 55) from low- and high-risk regions and have been characterized previously by MLST and for the presence of virulence factors (26, 27). Human gastric cancer (AGS) cells (CRL-1739; ATCC, Manassas, VA) were grown in phenol red-free Dulbeccos modied Eagle medium (DMEM) and Hams F12-K with 10% fetal bovine serum. Conuent AGS cells were infected at a multiplicity of infection (MOI) of 100 and incubated at 5% CO2 for the predetermined time. Microarray design. Seven annotated H. pylori genomes (26695 [NC_000915], J99 [NC_000921], HPAG1 [NC_008086 and NC_008087], B38 [NC_012973], P12 [NC_011498 and NC_011499], G27 [NC_011333 and NC_011334], and Shi470 [NC_010698]) were utilized to design probes using the Agilent eArray software (Agilent Technologies, Santa Clara, CA). Individual 60-bp oligonucleotide probes were designed for each gene within a genome. Additionally, 60-mer oligonucleotide probes were designed based on previously published sequences of H. pylori probes used by the Pathogen Functional Genomics Resource Center (pfgrc.jcvi.org). The accuracy of the probes was checked by using BLAST (28) prior to printing by Agilent. A total of 10,844 unique H. pylori genes, as dened by their Entrez IDs, were targeted. Multiple probes per Entrez ID were designed for genes in strains 26695 and J99 for a total of 14,987 probes per array. The ability of the microarray to detect enriched prokaryotic RNA was then tested. Total RNAs from AGS cells as well as from AGS cells infected with H. pylori strains ATCC 43504 or SS1 were processed using a Microbenrich kit (Applied Biosystems, Foster City, CA), which excluded eukaryotic mRNA and rRNA. RNA samples with and without prokaryotic RNA enrichment were then hybridized to the custom H. pylori microarrays (see Fig. S2 in the supplemental material). Minimal hybridization occurred from the eukaryotic samples, while hybridization was observed with H. pylori samples regardless of enrichment (see Fig. S2). Generating microarray data. Bacterial RNA for analysis was collected from three biological replicates per clinical isolate for a total of 18 samples. Following coculture of H. pylori and AGS cells for 1 h, RNAProtect Bacteria (Qiagen, Germantown, MD) was added to the culture media at a 2:1 ratio to prevent RNA degradation. AGS cells were scraped and allowed to incubate prior to collection by centrifugation at 5,000 g for 10 min (Sorvall RC5b plus). RNA was extracted following the RNeasy Mini Kit with RNase-free DNase treatment (Qiagen). Total prokaryotic and eukaryotic RNA was further processed using the Microbenrich kit, which

excluded eukaryotic mRNA and rRNA. The quality of the RNA was determined with the Agilent 2100 Bioanalyzer using an RNA 6000 Nano total RNA kit (Agilent Technologies) before and after the Microbenrich step to determine the removal of eukaryotic RNA. Total RNA was hybridized to the custom Agilent 8x15K H. pylori Microarrays following the One-Color Microarray-Based Gene expression Analysis, Low Input Quick Amp Labeling protocol. Briey, the method utilizes a T7 RNA polymerase that amplies RNA to cRNA while incorporating cyanine 3-labeled CTP. Cy-3 incorporation and cRNA levels were measured with a NanoDrop ND-1000 UV-Vis Spectrophotometer (NanoDrop Technologies, Inc., Wilmington, DE). Arrays were scanned using an Agilent Microarray Scanner, and data were extracted using Feature Extraction 9.1. Processing of raw microarray data. Processing of the raw gene expression data was performed using the limma library in R (29). Microarray data were rst corrected for background and then normalized by quantile using the NormalizeBetweenArrays function (30). Data were log2 transformed. Unsupervised clustering analysis was performed on the complete 14,987-probe data set using the R package pvclust (31). Genomic sequencing and selection of reduced probe set for expression analysis. The six isolates were grown on blood agar plates for 3 days, and the bacteria were resuspended in 200 l of phosphate-buffered saline (PBS). DNA was extracted using a Roche High Pure PCR template preparation kit, followed by RNase treatment (RiboShredder RNase Blend [Epicentre]) and cleanup with the High Pure PCR template preparation kit. Samples were submitted to the BioMicro Center at MIT for nextgeneration sequencing using an Illumina MiSeq system. Briey, samples were sonicated using Covaris S220 (Covaris Inc., Woburn, MA). Illumina libraries were constructed using the SPRI-TE robot (Beckman Coulter Inc., Danvers, MA) according to the manufacturers protocol. Samples were size selected for a 200- to 400-bp range using a BluePippin (Sage Science, Beverly, MA). Samples were quantied and pooled using the LC480 (Roche) and loaded on the Illumina MiSeq for sequencing. The 150-bp paired-end sequencing results generated by MiSeq from the six samples were aligned to the seven H. pylori reference genomes from which the gene expression probe sequences were derived using Bowtie2.0.6 (32). The aligned data were then used to perform variant calling using SAMtools mpileup (33). mpileup les were processed and ltered to generate high-quality variants (depth, 10; quality, 10; and homozygous for variant). The microarray probe sequences were then aligned to the reference genome from which they were designed using BLASTN (34). To determine the number of variants in each clinical isolate compared to the microarray probes, the high-quality variants that mapped to the probe regions were aggregated. Gene expression probes were selected if they had 0 or 1 variant compared within the 60-nucleotide probe sequence across the six samples. This probe set was further reduced by collapsing multiple expression values from probes with identical sequences to a single averaged expression value. Analysis of H. pylori mRNA expression. Signicance analysis was performed using signicance analysis of microarrays (SAM) (35) with the set of probes with 0 to 1 mismatch per probe across all six samples, and the groups were determined by the clustering analysis in a two-class unpaired analysis. SAM was implemented for stringent gene selection criteria using a d value, the test statistic used by SAM, of 1.32, which gave a falsediscovery rate (FDR) of 0.433% or 0.43 false positives for every 100 genes called. This FDR rate and a 1.5-fold change were used as cutoffs to generate the list of differentially expressed genes comparing the H. pylori strains. Differentially expressed genes that were assayed by multiple probes were reported as a single gene using the probe with the lowest q value. Fifteen genes were reported as upregulated in both groups by different probes and were removed from the analysis. Sets of differentially expressed genes were further classied using data from the COG classication of H. pylori (36), the eggNOG database (37), and previous publications (17, 23).

July 2013 Volume 81 Number 7

iai.asm.org 2469

Sheh et al.

IL-8 expression. Human gastric cancer (AGS) cells were seeded at 1 106/well in 6-well plates. Cells were cocultured with clinical strains for 6 h. Total RNA was isolated using TRIzol reagent, and 2 g of RNA from each sample was reverse transcribed as described previously (38). For IL-8 PCR, the primers used were as follows: sense, (5=-TAGCAAAATTGAGG CCAAGG-3=), and antisense, (5=-AAACCAAGGCACAGTGGAAC-3=). Real-time PCR was performed using SYBR green. One PCR cycle consisted of the following: 94C for 1 min, 62C for 1.5 min, and 72C for 1.8 min. Relative expression of IL-8 was determined using -actin as the internal control as described previously (38). Assessment of apoptosis. Apoptosis was assayed using an annexin V-uorescein isothiocyanate apoptosis detection kit (Oncogene Research Products, San Diego, CA) according to the manufacturers instructions and as described previously (39). Cells (2.5 106) were stained and acquired by ow cytometry using a Becton, Dickinson LSR II, and apoptosis was analyzed on the acquired cells with FlowJo software, as described previously (39). Motility analysis. Liquid cultures of the 6 Colombian isolates were resuspended in pH 7.0 brucella broth. Motility was monitored by live phase-contrast microscopy, and 300 frames were imaged with a Qimaging Qiclick video camera and Image-Pro Plus software. The resulting images were analyzed for bacterial movement using NIH ImageJ and the Difference Tracker plugin (40). Track data for each particle were further processed using Excel to compute mean speed and displacement. Bacteria had to be followed for at least eight consecutive frames to be considered in the analysis, and at least 300 tracks were analyzed per strain. Microarray data accession numbers. Raw and normalized data are available at the National Center for Biotechnology Information Gene Expression Omnibus (www.ncbi.nlm.nih.gov/geo/), series record GSE41497.

RESULTS

Clustering analysis classies H. pylori strains by phylogeographic origin and not geography. To explore the utility of gene expression proling in classifying H. pylori strains, three H. pylori strains from a Colombian region with high gastric cancer incidence (PZ5056, PZ5080, and PZ5086) and three strains from a Colombian region with low gastric cancer incidence (PZ5004, PZ5024, and PZ5026) were analyzed after infecting AGS cells for 1 h. Three separate infections were performed for each isolate (n 3 for each strain). Unsupervised clustering of the entire microarray data set for the 18 arrays correctly clustered the three replicate infections for each strain, demonstrating the reproducibility of the coculture system and subsequent processing of RNA (Fig. 1). The six strains were then grouped into two main clusters with two low-risk strains (PZ5004 and PZ5024) clustering together, while the remaining strains (three high-risk strains as well as one lowrisk strain [PZ5026]) clustered together (Fig. 1). The second cluster was further subdivided into two clusters, with two closely related high-risk strains (PZ5080 and PZ5086) in one group and a low-risk strain (PZ5026) and a high-risk strain (PZ5056) in the other. These clusters had scores of 99 in pvclust, implying strong support in the data set. It has been previously reported that PZ5026 and the three high-risk strains in the study have ancestral European origins, while PZ5004 and PZ5024 are of African origin (26). Therefore, unsupervised analysis of transcriptomes showed that there are distinct gene expression patterns associated with the phylogeographic origin of H. pylori strains. Analysis of gene expression proles for signicance. As hierarchical clustering demonstrated that strain origin was a key factor in gene expression, differences in expression between strains of African origin (n 2) and strains of European origin (n 4) were evaluated. Due to considerable variation in gene sequence and

FIG 1 Clustering analysis of 6 Colombian isolates. Microarray data from three strains from a low gastric cancer risk area (PZ5004, PZ5024, and PZ5026) and three strains from a high gastric cancer risk area (PZ5056, PZ5080, and PZ5086) were assessed in triplicate. The dendrograms edges are represented by numbers, and letters denote the P values for the edge (au/bp). Approximately unbiased P value (au) is based on multiscale bootstrap resampling, and bootstrap probability P value (bp) is based on normal bootstrap sampling. a, au & bp P value, 0.001; b, au & bp P value, 0.01; c, au & bp P value, 0.1.

gene content between H. pylori strains, we rst set out to quantify the similarity between the reference strains used to design the microarray and the strains being proled. Following next-generation sequencing of all six strains, we performed variant calling to compare the genomic sequence of the Colombian isolates to the probes designed from seven reference H. pylori genomes. A total of 4,126 unique probes were selected for gene expression analysis based on the established criteria of having 0 to 1 mismatch over the probe region across all six strains (see Table S1 in the supplemental material). These 4,126 unique probes measured expression of genes present in all strains using probes with 98.3% identity. Using SAM to identify genes with a 1.5-fold change cutoff at a 0.43% false-discovery rate, European strains had increased expression of 146 genes and decreased expression of 350 genes compared to their African counterparts. The complete list of differentially expressed genes is provided in Table S2 in the supplemental material. Of the 146 genes with increased expression in strains of European origin, 89 were genes of known function and the remaining 57 had unknown function. Of the 350 genes with decreased expression in the European strains, 163 had known function and 187 were of unknown function. Genes with known function are summarized in Fig. 2. Analysis of the differentially expressed genes revealed distinct patterns of host adaptation between the European and African strains. We observed differences in genes ascribed for virulence, motility, and adapting to the host environment. Virulence genes. Genes associated with virulence, adhesion, and transformation were analyzed to determine differences that might affect pathogenesis. European strains had signicantly

2470

iai.asm.org

Infection and Immunity

Origin Alters Transcription and Virulence of H. pylori

FIG 2 Functional analysis of genes with increased expression by strain origin: 89 of 146 genes with increased expression in European strains (gray) and 163 of
350 genes with increased expression in African strains (black) had assigned functions.

higher expression of babB, vacA, and cagA than did African strains, given a false-discovery rate of 0.43%. In addition, European strains also increased expression of two other components of the Cag-type IV secretion system (Cag-T4SS), cag4 and a regulator of VirB11 (HP1451) (4142), and a natural transformation gene, comH (43). In contrast, African strains increased expression of the genes of other known virulence factors (tumor necrosis factor [TNF-]-inducing protein [tip],
TABLE 1 Select genes with increased expression in European strainsa
Fold change in gene expression (Eur/Afr) 455.23 2.11 4.09 181.39 648.00 2.16 2.29 2.38 2.11 2.20 1.95 2.09 2.85 2.28 5.37 4.88 2.06

neutrophil activating protein[(napA], and a VacA-like protein). African strains also had increased expression of CagT4SS components (cagQ, cagX, tnpB, comB2, comB8-2, virB2-2, virB7-2, virB9, virB9-2, and virB10-1). Motility. Ten genes involved in motility and chemotaxis were differentially expressed between the two groups (Tables 1 and 2). European strains had increased expression of the agellar hook (gD), as well as the lament cap (iD) and a gene encoding a

Entrez ID 8208115 7009865 889799 900205 889201 6964369 900277 6963751 7009897 7010018 899547 7010060 6296551 890393 890075 899842 6297539

Gene name Aldo-keto reductase (HELPY_1166) ansB babB cag4 cagA comH dsbC gD iD mviN pqqE recR ruvC tlpC vacA VirB11-interacting protein (HP1451) VirB11-like protein (HPSH_04300)

COG FC C EJ N S S O N N R R L L NT N R NU

COG ID COG0667 COG0252 COG5651 proNOG56164 proNOG39791 COG2143 COG1843 COG1345 COG0728 COG0612 COG0353 COG0817 COG0840 COG5651 COG1847 COG0630

COG description Predicted oxidoreductases

L-asparaginase

Ontology Acid acclimation Energy metabolism Cell envelope Cellular processes Cellular processes Cellular processes Cellular processes Motility/chemotaxis Motility/chemotaxis Translation DNA metabolism DNA metabolism Motility/chemotaxis Cellular processes Cellular processes Cellular processes

PPE-repeat proteins

Thioredoxin-related protein Flagellar hook capping protein Flagellar capping protein Putative virulence factor Predicted Zn-dependent peptidases Recombinational DNA repair protein (RecF pathway) Holliday junction resolvasome, endonuclease subunit Methyl-accepting chemotaxis protein PPE-repeat proteins Predicted RNA-binding protein T4SS pathway, VirB11 components

a As calculated by SAM software. COG, cluster of orthologous groups; Eur/Afr, European/African; FC, functional categories, as described previously (36); T4SS, type IV secretion system.

July 2013 Volume 81 Number 7

iai.asm.org 2471

Sheh et al.

TABLE 2 Select genes with increased expression in African strainsa

Gene ID 7010039 7010646 8208413 900043 889138 889393 899094 889572 889872 899977 890192 889278 4099357 899333 889257 899445 890331 7009855 899404 889399 4098910 4097953 7010016 6963912 4099196 898902 889164 7010440 899290 889564 7011076 7011080 889322 7011082 Gene name ackA addB amiF cagQ cagX Chlorohydrolase (jhp0252) comB2 comB8-2 comL aA hB hB2 iH iS frdA frpB frxA homA katA luxS napA recO ruvA sodB tlpB tnpB tip ureG VacA-like protein (HP0610) virB10-1 virB2-2 virB7-2 virB9 virB9-2 Fold change (Eur/Afr) 0.362 0.333 0.437 0.089 0.075 0.636 0.376 0.119 0.237 0.634 0.085 0.149 0.097 0.161 0.127 0.393 0.053 0.224 0.364 0.536 0.523 0.521 0.576 0.644 0.398 0.01 0.043 0.629 0.396 0.474 0.185 0.666 0.234 0.312 COG FC C R R U FR U U R N NU S NU NUO C P C P T P S L P NT L OK N U COG ID COG0282 COG0457 COG0388 COG3504 COG0402 COG3504 COG3736 COG4105 COG1344 COG1377 COG2257 COG1317 COG1516 COG1053 COG1629 COG0778 NOG127861 COG0753 COG1854 COG0783 NOG12171 COG0632 COG0605 COG0840 COG0675 NOG135792 COG0378 COG5651 COG2948 COG descriptor Acetate kinase FOG: TPR repeat Predicted amidohydrolase T4SS, VirB9 comp. Cytosine deaminase & metal-dependent hydrolases T4SS, VirB9 comp. T4SS, VirB8 comp. DNA uptake lipoprotein Flagellin and hook-associated proteins Flagellar biosynthesis pathway FlhB Homolog of cytoplasmic domain of agellar protein FlhB Flagellar biosynthesis/T3SS protein Flagellin-specic chaperone FliS Succinate dehydrogenase/fumarate reductase, avoprotein subunit Outer membrane receptor proteins, mostly Fe transport Nitroreductase Catalase Involved in autoinducer AI2 synthesis DNA-binding ferritin-like protein/oxidation damage protectant Recombination protein RecO Holliday junction resolvasome, DNAbinding subunit Superoxide dismutase Methyl-accepting chemotaxis protein Transposase and inactivated derivatives Ni2-binding GTPase involved in urease and hydrogenase PPE-repeat proteins T4SS pathway-VirB10 comp. Ontology Energy metabolism DNA metabolism Acid acclimation Cellular processes Cellular processes Cellular processes Cellular processes Cellular processes Cellular processes Cell envelope Motility/chemotaxis Motility/chemotaxis Cell envelope Cell envelope Energy metabolism Transport Energy metabolism Cellular processes Hypothetical Transport DNA metabolism DNA metabolism Cellular processes Motility/chemotaxis Cellular processes Acid acclimation Cell envelope Cellular processes Cellular processes Cellular processes Cellular processes Cellular processes

U U

COG3504 COG3504

T4SS pathway, VirB9 comp. T4SS pathway, VirB9 comp.

a As calculated by SAM software. COG, cluster of orthologous groups; Eur/Afr, European/African; FC, functional categories, as described previously (36); T4SS, type IV secretion system.

methyl-accepting protein associated with chemotaxis (tlpC). African strains had increased expression of substructural subunits of the agellum (aA), a component of the type III secretion system (iH), and genes that aid in agellar export/secretion (iS, hB, and hB2). African strains also had increased transcription of tlpB, another chemotaxis-associated gene. The luxS gene, which helps regulate motility in H. pylori, was also differentially expressed in African strains(4446). Host-pathogen interactions. European and African strains also expressed different genes that might aid in adapting to the host environment. In the stomach, host-derived stresses could include changes in pH, exposure to reactive oxygen and nitrogen species (RONS), and DNA damage. In our coculture system, acid acclimation genes for formamidase (amiF) and a urease accessory

protein (ureG) had increased expression in African strains, while asparaginase (ansB) had higher expression in European strains. African strains also had higher expression of lepA, a gene that encodes GTP-binding membrane protein, and atpF, the FoF1-type ATP synthase gene; while European strains had higher expression of the aldo-keto reductase (HELPY_1166) gene. While not canonical acid acclimation genes, these three genes have been reported to be required for H. pylori growth under acidic conditions (47). European strains expressed DNA repair proteins (recR and ruvC) at a higher level, while African strains increased expression of genes that would protect against RONS, such as those encoding avodoxin (dA), superoxide dismutase (sodB), and catalase (katA), and that would mediate DNA repair (recO and ruvA). African strains also expressed more fumarate reductase (frdA),

2472

iai.asm.org

Infection and Immunity

Origin Alters Transcription and Virulence of H. pylori

TABLE 3 Analysis of H. pylori motilitya

H. pylori strain PZ5004 PZ5024 PZ5026 PZ5056 PZ5080 PZ5086
a

TABLE 4 cagA expression in coculture and patient data

Mean displacement (m) SD 23.5 21.2 7.6 9.4 5.3 5.7 10.4 8.1 5.2 6.4 4.1 5.7 Strain ID PZ5004 PZ5024 PZ5026 PZ5056 PZ5080 PZ5086 cagA expression in coculture systema 5.88 1.07 5.12 0.18 14.69 0.24 15.07 0.13 13.81 0.17 15.78 0.28 Region risk levelb Low Low Low High High High MLST typec hpAfrica hpAfrica hpEurope hpEurope hpEurope hpEurope Histopathology score of original patientd 2.33 2 5.25 4.5 3.5 4.6

Mean speed (m/s) SD 29.3 17.1 11.8 6.2 11.7 6.2 20.9 7.9 11.6 6.2 9.6 5.2

H. pylori strains grown in liquid culture were resuspended in pH 7.0 brucella broth and imaged. Bacterial movement was analyzed using particle tracking software in ImageJ. A minimum of 300 tracks were analyzed for each strain, with each track dened as one bacterium tracked in at least 8 consecutive frames.

which has been previously identied as necessary for colonization in vivo (48). Lack of correlation between bacterial movement and strain origin. To determine whether differences in strain origin and expression of motility genes would result in a phenotypic change in H. pylori motility, we conducted videomicroscopy of H. pylori motility when the strains were exposed to pH 7.0 brucella broth. Analysis of the six strains showed that all six strains were motile. Average speed and displacement were calculated for each strain and were found to be independent of strain origin (Table 3). Correlation of cagA in coculture with histopathological ndings. Due to the association of CagA with gastric carcinoma, cagA expression levels in the coculture system were compared to previously published data on the histologic lesions in the patients from which the clinical isolates were derived(2627). Higher levels of cagA mRNA in the coculture system were observed in strains isolated from patients with increased histopathological scores (Table 4 and Fig. 3), and the correlation for these data sets is 0.83 (Spearman), indicating a strong positive relationship. IL-8 expression. Due to increased expression of the virulence factor cagA in the European strains, IL-8 expression was assessed in gastric epithelial cells following infection with ve of the six strains used in the microarray experiment. As shown in Fig. 4A, IL-8 mRNA levels were signicantly higher in the two European strains tested from the high-risk region, with increases of (20.1 5.8)-fold and (18.5 7.0)-fold for PZ5056 and PZ5086, respectively (P 0.001 for each versus control). In addition, the low-risk strain of European origin, PZ5026, also induced a signicant increase in IL-8 levels of (9.9 2.7)-fold versus control (P 0.01). In contrast, the low-risk African strains, PZ5004 and PZ5024, failed to induce a signicant increase in IL-8 mRNA expression (Fig. 4A). Additionally, the two high-risk European strains each induced a signicant increase in IL-8 levels compared to both of the low-risk African strains (P 0.001 for each), and the low-risk European strain also induced a signicant increase compared to the low-risk African strains (P 0.05 for each; Fig. 4A). Apoptosis. Because alterations in apoptosis have been linked to gastric cancer risk (49), the ability of ve of these clinical isolates to induce apoptosis in gastric epithelial cells was also assessed. As shown in Fig. 4B, the low-risk African strains PZ5004 and PZ5024 both induced a signicant increase in apoptosis, from 3.3% 0.7% in the uninfected control cells to 22.8% 4.8% and 17.8% 1.2%, respectively (P 0.01 versus control for each). In contrast, the two high-risk European strains, PZ5056 and PZ8086, caused only a modest increase in apoptosis that was not signicant, and

a cagA expression was analyzed by microarray using the probe designed against cagA for H. pylori J99 (889201). For each strain, the cagA signal was calculated for three independent biological samples, and the average and SD of the normalized and logged values are presented. The Spearman correlation is 0.83, indicating a positive relationship between cagA expression and the histopathological score. b H. pylori strains were isolated from regions of Colombia with either high or low risk for the development of gastric cancer. c MLST analysis (26) was previously used to characterize the ancestral origin of the H. pylori strains. d Previously published histopathology scores describing the severity of lesions observed in antral gastric tissue from which the H. pylori strains were isolated (26, 27).

they both exhibited less induction of apoptosis than did the lowrisk African strains. The low-risk European strain showed an intermediate phenotype in this assay, as there was a signicant increase in apoptosis to 13.0% 4.3% (P 0.05 versus control).
DISCUSSION

Helicobacter pylori has infected human hosts for many millennia (50). Due to the plasticity of its genome and its capacity to acquire genes from bacterial donors in the environment, H. pylori has developed an extensive array of defense mechanisms and virulence factors that permit its survival in the harsh gastric environment (8, 51). However, environmental pressures, such as diet or coinfection with parasites, may have driven H. pylori strains on divergent evolutionary paths regarding virulence. Less virulent or less interactive strains might colonize with populations distributed toward the lumen and elicit mild immune responses with commensurate development of mild gastritis, while more virulent or more interactive strains may colonize closer to the epithelium and increase the impact on host tissues and gastric cancer risk (22, 52). In this study, unsupervised clustering of gene expression data demonstrated that gene expression signatures could effectively distinguish the more virulent strains of European origin and the less virulent strains of African origin. This classication corroborates previous MLST data, which classied the 2 low-risk strains (PZ5004 and PZ5024) as having African origins, while the remaining strains, including PZ5026 (a strain isolated from an individual at a low-risk locale), had European origins (26). Gene expression analysis of H. pylori isolates in this study provided a useful tool to classify the pathogenicity of H. pylori strains. While genotyping for vacA s1m1 and cagA status is useful to assess the risk of developing duodenal ulcers or gastric cancer (1), the role of ancestral origin on the modulation of these virulence factors has not been studied extensively. All strains used in this study were cagA and vacA s1m1 but exhibited signicant expression differences in virulence factors. European strains had increased expression of cagA, vacA, babB, and cag4 compared to African strains. babB encodes an adhesin that mediates site-specic colonization and increases gastric pathology (53). cag4 is thought to encode a lytic transglycosylase, homologous to VirB1,

July 2013 Volume 81 Number 7

iai.asm.org 2473

Sheh et al.

FIG 3 Histopathology in patients from which PZ5004 (A, B) (nonatrophic gastritis), PZ5056 (C, D) (intestinal metaplasia), and PZ5086 (E, F) (intestinal metaplasia) were isolated. PZ5004 shows diffuse mononuclear leukocytic inltration throughout and well-preserved glands (arrow). PZ5056 and PZ5086 both show irregular Goblet cells (arrow). Images are at magnications of 100 (left) and 200 (right). Images courtesy of Maria Blanca Piazuelo.

which may degrade the bacterial wall to allow the Cag-T4SS to form (54). CagA requires direct contact to the epithelium to mediate structural changes (15). Increased cagA and babB expression may reect closer contact between European strains and the epithelium. Interestingly, two well-characterized virulence genes showing increased expression in African strains (tip and napA) both encode secreted factors (5556), which may suggest colonization further away from the epithelium. Further studies will be necessary to elucidate the dynamics of attachment to host cells between different Colombian strains of H. pylori. Gene expression changes in other bacterial functions necessary for colonization (acid acclimation, response to RONS, motility, and DNA repair) were observed. To evaluate the signicance of expression changes in motility and chemotaxis genes, we evalu-

ated H. pylori motility in pH 7 culture media. While all six clinical isolates were motile, analysis of H. pylori mean speed and displacement showed no correlation between swimming behavior and strain origin. Further studies measuring the differences between African and European strains upon attachment to epithelial cells, acid exposure, or RONS exposure are required to determine which of these phenotypes are linked to phylogeographic origin. We hypothesize that the phenotypes associated with strain origin are not necessary for successful infection by H. pylori. While cagA and vacA are strongly linked to strain origin (57), they are not necessary for infection or survival (1). Motility is an indispensable function for H. pylori infection (4), so natural selection may limit the range of motility phenotypes that result from evolutionary or environmental pressures.

2474

iai.asm.org

Infection and Immunity

Origin Alters Transcription and Virulence of H. pylori

FIG 4 Levels of IL-8 mRNA expression (A) and apoptosis (B) in AGS cells stimulated with H. pylori clinical isolates. Cells were cocultured with H. pylori strains at a multiplicity of infection of 200 for 6 h (A) and for 24 h (B). *, P 0.05; **, P 0.01; ***, P 0.001 versus unstimulated control cells; , P 0.05; , P 0.001 compared to both low-risk African strains (PZ5004 and PZ5024); #, P 0.05 versus African strain PZ5004.

The CXC chemokine family, which includes IL-8 and CXCL1, consists of proinammatory cytokines that are released in response to H. pylori infection. These cytokines recruit neutrophils and macrophages to sites of inammation, but prolonged exposure to these cytokines has been linked clinically to increased gastric cancer risk (5863). H. pylori infection has been associated with increased IL-8 expression and decreased p27 levels (64). Both reduced p27 levels and increased IL-8 levels have been shown to mediate decreased apoptosis (6568). When exposed to morevirulent European strains expressing higher levels of cagA, gastric epithelial cells increased IL-8 expression, which promotes the activation and inltration of inammatory cells and the reduction of apoptosis. This combination may make European strains especially harmful, as it actively promotes the formation of a tumor microenvironment and allows the accumulation of host DNA damage by avoiding programmed cell death. In conclusion, gene expression analysis of the six H. pylori isolates provided a view of how African and European strains use divergent yet successful virulence strategies to persistently infect their hosts (summarized in Table 5). Our results demonstrate that expression differences in H. pylori strains of African and European origins can be associated with phenotypic changes associated with gastric pathogenesis. These transcriptional changes may reect differences in niches within the mucosa of more or less virulent H. pylori strains (8, 52). Furthermore, the more robust inammatory response to European strains, coupled with a decrease in apoptosis, as well as increased expression of cagA, vacA, and babB, may provide mechanistic insights into the increase in DNA damage, the more-severe histopathologic gastric lesions in humans, and

the overall increased gastric cancer risk in subjects infected with European strains compared to subjects infected with African strains (26). A better understanding of how host-pathogen interactions vary between different H. pylori strains and how these changes modulate gastric disease severity will improve our ability to manage chronic inammatory conditions.
ACKNOWLEDGMENTS
We thank Maria Blanca Piazuelo for providing histopathology images from patients from which H. pylori strains were isolated. This work was supported by National Institutes of Health grants P01CA028842 (J.G.F., K.T.W., and P.C.), P01CA026731 (J.G.F.), P30ES002109 (J.G.F.), R01DK053620 (K.T.W.), P01CA116087 (K.T.W.), UL1RR024975 (Vanderbilt CTSA, Pilot Project to K.T.W.), the Flow Cytometry Core of the Vanderbilt University Digestive Disease Research Center grant (P30DK058404), and a Merit Review Grant from the Ofce of Medical Research, Department of Veterans Affairs (K.T.W.).

REFERENCES
1. Kusters JG, van Vliet AH, Kuipers EJ. 2006. Pathogenesis of Helicobacter pylori infection. Clin. Microbiol. Rev. 19:449 490. 2. Williams SM, Chen YT, Andermann TM, Carter JE, McGee DJ, Ottemann KM. 2007. Helicobacter pylori chemotaxis modulates inammation and bacterium-gastric epithelium interactions in infected mice. Infect. Immun. 75:37473757. 3. Croxen MA, Sisson G, Melano R, Hoffman PS. 2006. The Helicobacter pylori chemotaxis receptor TlpB (HP0103) is required for pH taxis and for colonization of the gastric mucosa. J. Bacteriol. 188:2656 2665. 4. Ottemann KM, Lowenthal AC. 2002. Helicobacter pylori uses motility for initial colonization and to attain robust infection. Infect. Immun. 70: 1984 1990. 5. Bauerfeind P, Garner R, Dunn BE, Mobley HL. 1997. Synthesis and activity of Helicobacter pylori urease and catalase at low pH. Gut 40: 2530. 6. Wen Y, Feng J, Scott DR, Marcus EA, Sachs G. 2007. The HP0165HP0166 two-component system (ArsRS) regulates acid-induced expression of HP1186 alpha-carbonic anhydrase in Helicobacter pylori by activating the pH-dependent promoter. J. Bacteriol. 189:2426 2434. 7. Gerhard M, Lehn N, Neumayer N, Boren T, Rad R, Schepp W, Miehlke S, Classen M, Prinz C. 1999. Clinical relevance of the Helicobacter pylori gene for blood-group antigen-binding adhesin. Proc. Natl. Acad. Sci. U. S. A. 96:12778 12783. 8. Tan S, Tompkins LS, Amieva MR. 2009. Helicobacter pylori usurps cell polarity to turn the cell surface into a replicative niche. PLoS Pathog. 5:e1000407. doi:10.1371/journal.ppat.1000407.

TABLE 5 Summary of key differences between African and European H. pylori strains
Characteristic cagA and vacA expression Motility (speed and displacement) Expression of IL-8 by AGS cells Induction of apoptosis in AGS cells Histopathology African origin Low No correlation Low High Low European origin High No correlation High Low High

July 2013 Volume 81 Number 7

iai.asm.org 2475

Sheh et al.

9. Gonzalez CA, Figueiredo C, Lic CB, Ferreira RM, Pardo ML, Ruiz Liso JM, Alonso P, Sala N, Capella G, Sanz-Anquela JM. 2011. Helicobacter pylori cagA and vacA genotypes as predictors of progression of gastric preneoplastic lesions: a long-term follow-up in a high-risk area in Spain. Am. J. Gastroenterol. 106:867 874. 10. Leunk RD, Johnson PT, David BC, Kraft WG, Morgan DR. 1988. Cytotoxic activity in broth-culture ltrates of Campylobacter pylori. J. Med. Microbiol. 26:9399. 11. Covacci A, Censini S, Bugnoli M, Petracca R, Burroni D, Macchia G, Massone A, Papini E, Xiang Z, Figura N, Rappuoli R. 1993. Molecular characterization of the 128-kDa immunodominant antigen of Helicobacter pylori associated with cytotoxicity and duodenal ulcer. Proc. Natl. Acad. Sci. U. S. A. 90:57915795. 12. Hatakeyama M. 2008. SagA of CagA in Helicobacter pylori pathogenesis. Curr. Opin. Microbiol. 11:30 37. 13. Yamaoka Y, Kudo T, Lu H, Casola A, Brasier AR, Graham DY. 2004. Role of interferon-stimulated responsive element-like element in interleukin-8 promoter in Helicobacter pylori infection. Gastroenterology 126: 1030 1043. 14. Tabassam FH, Graham DY, Yamaoka Y. 2009. Helicobacter pylori activate epidermal growth factor receptor- and phosphatidylinositol 3-OH kinase-dependent Akt and glycogen synthase kinase 3beta phosphorylation. Cell. Microbiol. 11:70 82. 15. Hatakeyama M. 2006. Helicobacter pylori CagAa bacterial intruder conspiring gastric carcinogenesis. Int. J. Cancer 119:12171223. 16. Crabtree JE, Ferrero RL, Kusters JG. 2002. The mouse colonizing Helicobacter pylori strain SS1 may lack a functional cag pathogenicity island. Helicobacter 7:139 140, author reply 140 141. 17. Merrell DS, Goodrich ML, Otto G, Tompkins LS, Falkow S. 2003. pH-regulated gene expression of the gastric pathogen Helicobacter pylori. Infect. Immun. 71:3529 3539. 18. Merrell DS, Thompson LJ, Kim CC, Mitchell H, Tompkins LS, Lee A, Falkow S. 2003. Growth phase-dependent response of Helicobacter pylori to iron starvation. Infect. Immun. 71:6510 6525. 19. Thompson LJ, Merrell DS, Neilan BA, Mitchell H, Lee A, Falkow S. 2003. Gene expression proling of Helicobacter pylori reveals a growthphase-dependent switch in virulence gene expression. Infect. Immun. 71: 26432655. 20. Scott DR, Marcus EA, Wen Y, Oh J, Sachs G. 2007. Gene expression in vivo shows that Helicobacter pylori colonizes an acidic niche on the gastric surface. Proc. Natl. Acad. Sci. U. S. A. 104:72357240. 21. Gancz H, Censini S, Merrell DS. 2006. Iron and pH homeostasis intersect at the level of Fur regulation in the gastric pathogen Helicobacter pylori. Infect. Immun. 74:602 614. 22. Guillemin K, Salama NR, Tompkins LS, Falkow S. 2002. Cag pathogenicity island-specic responses of gastric epithelial cells to Helicobacter pylori infection. Proc. Natl. Acad. Sci. U. S. A. 99:15136 15141. 23. Kim N, Marcus EA, Wen Y, Weeks DL, Scott DR, Jung HC, Song IS, Sachs G. 2004. Genes of Helicobacter pylori regulated by attachment to AGS cells. Infect. Immun. 72:2358 2368. 24. Bravo LE, van Doom LJ, Realpe JL, Correa P. 2002. Virulence-associated genotypes of Helicobacter pylori: do they explain the African enigma? Am. J. Gastroenterol. 97:2839 2842. 25. Whary MT, Sundina N, Bravo LE, Correa P, Quinones F, Caro F, Fox JG. 2005. Intestinal helminthiasis in Colombian children promotes a Th2 response to Helicobacter pylori: possible implications for gastric carcinogenesis. Cancer Epidemiol. Biomarkers Prev. 14:1464 1469. 26. de Sablet T, Piazuelo MB, Shaffer CL, Schneider BG, Asim M, Chaturvedi R, Bravo LE, Sicinschi LA, Delgado AG, Mera RM, Israel DA, Romero-Gallo J, Peek RM, Jr, Cover TL, Correa P, Wilson KT. 2011. Phylogeographic origin of Helicobacter pylori is a determinant of gastric cancer risk. Gut 60:1189 1195. 27. Loh JT, Shaffer CL, Piazuelo MB, Bravo LE, McClain MS, Correa P, Cover TL. 2011. Analysis of cagA in Helicobacter pylori strains from Colombian populations with contrasting gastric cancer risk reveals a biomarker for disease severity. Cancer Epidemiol. Biomarkers Prev. 20:2237 2249. 28. Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389 3402. 29. Smyth GK. 2004. Linear models and empirical Bayes methods for assessing differential expression in microarray experiments. Stat Appl. Genet. Mol. Biol. 3:Article 3.

30. R Development Core Team. 2008. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. 31. Suzuki R, Shimodaira H. 2006. Pvclust: an R package for assessing the uncertainty in hierarchical clustering. Bioinformatics 22:1540 1542. 32. Langmead B, Salzberg SL. 2012. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9:357359. 33. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R. 2009. The sequence alignment/map format and SAMtools. Bioinformatics 25:2078 2079. 34. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403 410. 35. Tusher VG, Tibshirani R, Chu G. 2001. Signicance analysis of microarrays applied to the ionizing radiation response. Proc. Natl. Acad. Sci. U. S. A. 98:5116 5121. 36. Tatusov RL, Fedorova ND, Jackson JD, Jacobs AR, Kiryutin B, Koonin EV, Krylov DM, Mazumder R, Mekhedov SL, Nikolskaya AN, Rao BS, Smirnov S, Sverdlov AV, Vasudevan S, Wolf YI, Yin JJ, Natale DA. 2003. The COG database: an updated version includes eukaryotes. BMC Bioinformatics 4:41. doi:10.1186/1471-2105-4-41. 37. Powell S, Szklarczyk D, Trachana K, Roth A, Kuhn M, Muller J, Arnold R, Rattei T, Letunic I, Doerks T, Jensen LJ, von Mering C, Bork P. 2012. eggNOG v3.0: orthologous groups covering 1133 organisms at 41 different taxonomic ranges. Nucleic Acids Res. 40:D284 D289. 38. Chaturvedi R, Asim M, Hoge S, Lewis ND, Singh K, Barry DP, de Sablet T, Piazuelo MB, Sarvaria AR, Cheng Y, Closs EI, Casero RA, Jr, Gobert AP, Wilson KT. 2010. Polyamines impair immunity to Helicobacter pylori by inhibiting L-arginine uptake required for nitric oxide production. Gastroenterology 139:1686 1698. 39. Chaturvedi R, Asim M, Romero-Gallo J, Barry DP, Hoge S, de Sablet T, Delgado AG, Wroblewski LE, Piazuelo MB, Yan F, Israel DA, Casero RA, Jr, Correa P, Gobert AP, Polk DB, Peek RM, Jr, Wilson KT. 2011. Spermine oxidase mediates the gastric cancer risk associated with Helicobacter pylori CagA. Gastroenterology 141:1696 1708. 40. Andrews S, Gilley J, Coleman MP. 2010. Difference Tracker: ImageJ plugins for fully automated analysis of multiple axonal transport parameters. J. Neurosci. Methods 193:281287. 41. Hare S, Fischer W, Williams R, Terradot L, Bayliss R, Haas R, Waksman G. 2007. Identication, structure and mode of action of a new regulator of the Helicobacter pylori HP0525 ATPase. EMBO J. 26:4926 4934. 42. Terradot L, Waksman G. 2011. Architecture of the Helicobacter pylori Cag-type IV secretion system. FEBS J. 278:12131222. 43. Smeets LC, Bijlsma JJE, Boomkens SY, Vandenbroucke-Grauls CMJE, Kusters JG. 2000. comH, a novel gene essential for natural transformation of Helicobacter pylori. J. Bacteriol. 182:3948 3954. 44. Rader BA, Campagna SR, Semmelhack MF, Bassler BL, Guillemin K. 2007. The quorum-sensing molecule autoinducer 2 regulates motility and agellar morphogenesis in Helicobacter pylori. J. Bacteriol. 189:6109 6117. 45. Osaki T, Hanawa T, Manzoku T, Fukuda M, Kawakami H, Suzuki H, Yamaguchi H, Yan X, Taguchi H, Kurata S, Kamiya S. 2006. Mutation of luxS affects motility and infectivity of Helicobacter pylori in gastric mucosa of a Mongolian gerbil model. J. Med. Microbiol. 55:14771485. 46. Shen F, Hobley L, Doherty N, Loh JT, Cover TL, Sockett RE, Hardie KR, Atherton JC. 2010. In Helicobacter pylori auto-inducer-2, but not LuxS/MccAB catalysed reverse transsulphuration, regulates motility through modulation of agellar gene transcription. BMC Microbiol. 10: 210. doi:10.1186/1471-2180-10-210. 47. Bijlsma JJE, Lie Ling A-M, Nootenboom IC, Vandenbroucke-Grauls CMJE, Kusters JG. 2000. Identication of loci essential for the growth of Helicobacter pylori under acidic conditions. J. Infect. Dis. 182:1566 1569. 48. Ge Z, Feng Y, Dangler CA, Xu S, Taylor NS, Fox JG. 2000. Fumarate reductase is essential for Helicobacter pylori colonization of the mouse stomach. Microb. Pathog. 29:279 287. 49. Galysh R, Powell SM. 2009. Cancer genetics of human gastric adenocarcinoma, p 251283. In Wang TC, Fox JG, Giraud AS (ed), The biology of gastric cancers. Springer, New York, New York, USA. 50. Linz B, Balloux F, Moodley Y, Manica A, Liu H, Roumagnac P, Falush D, Stamer C, Prugnolle F, van der Merwe SW, Yamaoka Y, Graham DY, Perez-Trallero E, Wadstrom T, Suerbaum S, Achtman M. 2007. An African origin for the intimate association between humans and Helicobacter pylori. Nature 445:915918.

2476

iai.asm.org

Infection and Immunity

Origin Alters Transcription and Virulence of H. pylori

51. Ahmed N, Tenguria S, Nandanwar N. 2009. Helicobacter pyloria seasoned pathogen by any other name. Gut Pathog. 1:24. doi:10.1186 /1757-4749-1-24. 52. Blaser MJ. 1998. Science, medicine, and the future: Helicobacter pylori and gastric diseases. Br. Med. J. 316:1507. 53. Matteo MJ, Armitano RI, Romeo M, Wonaga A, Olmos M, Catalano M. 2011. Helicobacter pylori bab genes during chronic colonization. Int. J. Mol. Epidemiol. Genet. 2:286 291. 54. Zhong Q, Shao S, Mu R, Wang H, Huang S, Han J, Huang H, Tian S. 2011. Characterization of peptidoglycan hydrolase in Cag pathogenicity island of Helicobacter pylori. Mol. Biol. Rep. 38:503509. 55. Suganuma M, Kuzuhara T, Yamaguchi K, Fujiki H. 2006. Carcinogenic role of tumor necrosis factor-alpha inducing protein of Helicobacter pylori in human stomach. J. Biochem. Mol. Biol. 39:1. 56. Evans D, Evans DG, Takemura T, Nakano H, Lampert HC, Graham DY, Granger DN, Kvietys PR. 1995. Characterization of a Helicobacter pylori neutrophil-activating protein. Infect. Immun. 63:22132220. 57. Duncan SS, Valk PL, McClain MS, Shaffer CL, Metcalf JA, Bordenstein SR, Cover TL. 2013. Comparative genomic analysis of East Asian and non-Asian Helicobacter pylori strains identies rapidly evolving genes. PLoS One 8:e55120. doi:10.1371/journal.pone.0055120. 58. Sheh A, Ge Z, Parry NM, Muthupalani S, Rager JE, Raczynski AR, Mobley MW, McCabe AF, Fry RC, Wang TC, Fox JG. 2011. 17betaestradiol and tamoxifen prevent gastric cancer by modulating leukocyte recruitment and oncogenic pathways in Helicobacter pylori-infected INS-GAS male mice. Cancer Prev. Res. (Phila.) 4:1426 1435. 59. Okumura T, Ericksen RE, Takaishi S, Wang SS, Dubeykovskiy Z, Shibata W, Betz KS, Muthupalani S, Rogers AB, Fox JG, Rustgi AK, Wang TC. 2010. K-ras mutation targeted to gastric tissue progenitor cells results in chronic inammation, an altered microenvironment, and progression to intraepithelial neoplasia. Cancer Res. 70:8435 8445. 60. Vandercappellen J, Van Damme J, Struyf S. 2008. The role of CXC chemokines and their receptors in cancer. Cancer Lett. 267:226 244.

61. Jung JJ, Noh S, Jeung HC, Jung M, Kim TS, Noh SH, Roh JK, Chung HC, Rha SY. 2010. Chemokine growth-regulated oncogene 1 as a putative biomarker for gastric cancer progression. Cancer Sci. 101: 2200 2206. 62. Junnila S, Kokkola A, Mizuguchi T, Hirata K, Karjalainen-Lindsberg ML, Puolakkainen P, Monni O. 2010. Gene expression analysis identies over-expression of CXCL1, SPARC, SPP1, and SULF1 in gastric cancer. Genes Chromosomes Cancer 49:28 39. 63. Taguchi A, Ohmiya N, Shirai K, Mabuchi N, Itoh A, Hirooka Y, Niwa Y, Goto H. 2005. Interleukin-8 promoter polymorphism increases the risk of atrophic gastritis and gastric cancer in Japan. Cancer Epidemiol. Biomarkers Prev. 14:24872493. 64. Kim SS, Meitner P, Konkin TA, Cho YS, Resnick MB, Moss SF. 2006. Altered expression of Skp2, c-Myc and p27 proteins but not mRNA after H. pylori eradication in chronic gastritis. Mod. Pathol. 19:49 58. 65. Shirin H, Sordillo EM, Kolevska TK, Hibshoosh H, Kawabata Y, Oh SH, Kuebler JF, Delohery T, Weghorst CM, Weinstein IB, Moss SF. 2000. Chronic Helicobacter pylori infection induces an apoptosisresistant phenotype associated with decreased expression of p27(kip1). Infect. Immun. 68:53215328. 66. Kuzushita N, Rogers AB, Monti NA, Whary MT, Park MJ, Aswad BI, Shirin H, Koff A, Eguchi H, Moss SF. 2005. p27kip1 deciency confers susceptibility to gastric carcinogenesis in Helicobacter pylori-infected mice. Gastroenterology 129:1544 1556. 67. Abdollahi T, Robertson NM, Abdollahi A, Litwack G. 2003. Identication of interleukin 8 as an inhibitor of tumor necrosis factor-related apoptosis-inducing ligand-induced apoptosis in the ovarian carcinoma cell line OVCAR3. Cancer Res. 63:4521 4526. 68. Wilson C, Wilson T, Johnston PG, Longley DB, Waugh DJ. 2008. Interleukin-8 signaling attenuates TRAIL- and chemotherapy-induced apoptosis through transcriptional regulation of c-FLIP in prostate cancer cells. Mol. Cancer Ther. 7:2649 2661.

July 2013 Volume 81 Number 7

iai.asm.org 2477